Funded by NIA since 1989, the Epidemiology of Dementia Study has focused on identifying incident cases of dementia and determining the risk factors associated with clinical dementia, particularly Alzheimer's disease (AD) in the large prospective community-based Framingham Study. Beginning with the establishment of a dementia-free cohort in 1976-78, this study has documented >600 subjects with incident dementia, largely from the Original cohort recruited in 1948. In recent years, with the aging of the children of the Original cohort (and their spouses), a number of these Generation 2 subjects have been documented to have cognitive decline, and MRI markers of brain injury as well as Mild Cognitive Impairment (MCI) and dementia. We have provided community-based estimates of incidence and lifetime risk of AD, the risk of dementia following stroke, and have identified a number of lifestyle, vascular, novel biomarker and more recently genetic risk factors for AD. Through an ongoing brain donation program established in 1995, we have performed detailed neuropathologic examination of approximately 140 brains. In this application we will focus on identifying persons with subclinical disease by intensifying surveillance to identify subjects with probable MCI and will follow them with annual assessments. We will relate a wealth of previously collected risk factors to the risk of incident AD and all-dementia, to AD endophenotypes, and to the risk of conversion from MCI to clinical AD. These risk factors include genetic (genome-wide, candidate gene and gene expression) and environmental (lifestyle, vascular, metabolic measures gathered repeatedly since midlife) data, circulating biomarker levels and baseline imaging data. To this array of putative risk markers we propose to add: (1) a key biomarker, circulating beta-amyloid (A240 and A242) levels; (2) a third round of quantitative brain MRI and NP testing to more precisely document change, and distinguish those with consistent signs of brain atrophy and cognitive decline at highest risk of transitioning to MCI and/or dementia; and, (3) detailed post-processing MR analyses of additional regions of interest from prior and newly acquired MRI images to enhance detection of subtle changes in brain morphology over time. The goal is to incorporate these new and existing data into a risk prediction score for AD and dementia and validate it in an independent epidemiological cohort. The aim of this risk profile is to identify persons at highest risk for developing AD, thus permitting targeted preventive and therapeutic interventions, including enrollment of susceptible subjects for clinical trials of promising therapies.